Adaptive CCLM Parameter Selection for Chroma Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face inefficiencies in intra prediction and motion vector prediction, particularly in handling chroma components, leading to suboptimal compression ratios and increased data requirements.
Innovation Solution
The implementation of a cross-component linear model (CCLM) for chroma block reconstruction, where adjustment parameters are determined based on collocated luma block samples to enhance prediction accuracy and reduce redundancy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traditional intra coding is used for chroma block reconstruction, then device complexity is reduced, but compression efficiency deteriorates
Solution Approach 1:
The patent applies parameter changes by introducing adjustable offset parameters and adjustment parameters that modify the CCLM prediction model. These parameters are adaptively selected based on picture type (I, P, B frames) and block characteristics, allowing the system to optimize compression efficiency without requiring complex additional processing structures.
Solution Approach 2:
The patent implements dynamics by making the offset parameter selection adaptive rather than static. The system dynamically selects between different offset parameter determination methods based on the current picture type and block characteristics, enabling the coding system to adapt to varying content requirements while maintaining reasonable complexity.
2Productivity
If cross-component linear model with adjustment parameters is used for chroma block reconstruction, then compression efficiency is improved, but device complexity increases
Solution Approach 1:
The patent uses parameter changes by introducing adjustment parameters that modify the offset values in the CCLM model. These parameters are derived from luma block samples and applied to chroma blocks, improving prediction accuracy. The parameters are calculated using straightforward arithmetic operations rather than complex algorithms, balancing efficiency improvement with acceptable complexity increase.
Solution Approach 2:
The patent applies segmentation by dividing the chroma block into multiple regions and applying different offset parameters to different regions. This allows localized optimization of prediction accuracy without requiring complex processing for the entire block, thereby improving compression efficiency while controlling the increase in coding complexity.
3Measurement precision
If more samples from collocated luma block are used for parameter determination, then prediction accuracy is improved, but data processing requirements increase
Solution Approach 1:
The patent applies local quality by selectively using samples from different regions of the collocated luma block based on the chroma block position. For example, top-left chroma blocks use top-left luma samples, while bottom-right chroma blocks use bottom-right luma samples. This localized approach improves prediction accuracy for each specific region without requiring processing of all luma samples, thus controlling data processing requirements.
Solution Approach 2:
The patent segments the luma block samples into different groups based on their spatial relationship to the chroma block. By dividing the sample selection into specific regions and assigning appropriate samples to corresponding chroma regions, the system achieves high prediction accuracy while limiting the total number of samples that need to be processed.
Data Source
AI summary
A method and an apparatus including processing circuitry for video processing are provided. For a first region in a chroma block in a current picture, the processing circuitry is configured to update an offset parameter in a cross-component linear model (CCLM) mode based on a first subset of reconstructed samples in a luma block that is collocated with the chroma block in the current picture. The first subset of the reconstructed samples does not include one or more reconstructed samples in the luma block. The processing circuitry is configured to update a slope parameter in the CCLM mode based on a second subset of the reconstructed samples in the luma block and encode the first region in the chroma block based on the updated offset parameter and the updated slope parameter using the CCLM mode.